OUTPUT of run 7 of at most 100: ``` MLP NEURON NORM QUANTILES: input row / output column 0 [0.3266 0.3623 0.3895 0.4232 0.9669 2.6887 4.4219] [0.1318 0.1623 0.194 0.2373 1.2717 2.154 4.7954] 1 [0.3913 1.7931 1.9056 2.004 2.1051 2.2776 3.3711] [0.1986 1.7341 1.8466 1.9219 2.0072 2.1576 2.6756] 2 [1.7221 1.9046 1.9707 2.0103 2.0551 2.1253 2.4825] [1.5505 1.7804 1.8885 1.9574 2.0251 2.1289 2.4465] 3 [1.7763 1.8813 1.9381 1.9761 2.014 2.0986 3.1298] [1.4454 1.735 1.9208 2.0615 2.2221 2.491 4.8197] LAST MLP OUTPUT-COLUMN GEOMETRY nearest cosine percentiles: [0.459 0.5382 0.5696 0.605 0.6485 0.7103 0.9073 0.9462] k 20 inertia 1139.135 cluster sizes [107 96 89 88 87 85 84 83 81 78 74 74 73 71 69 68 66 64 52 47] centroid norms [0.7 0.475 0.529 0.49 0.448 0.475 0.5 0.508 0.465 0.497 0.502 0.469 0.499 0.491 0.481 0.491 0.505 0.471 0.481 0.57 ] k 24 inertia 1117.005 cluster sizes [93 79 77 74 73 72 70 67 66 66 66 66 65 58 58 58 57 57 57 56 54 51 48 48] centroid norms [0.6 0.726 0.511 0.435 0.501 0.508 0.458 0.507 0.485 0.528 0.5 0.564 0.485 0.506 0.488 0.5 0.507 0.477 0.486 0.513 0.523 0.517 0.519 0.54 ] k 28 inertia 1095.936 cluster sizes [78 76 73 68 63 63 62 61 59 58 58 55 55 55 53 52 51 50 49 48 48 47 46 44 43 43 40 38] centroid norms [0.725 0.653 0.504 0.506 0.511 0.505 0.533 0.588 0.479 0.508 0.488 0.496 0.5 0.524 0.493 0.548 0.456 0.556 0.499 0.517 0.532 0.522 0.492 0.508 0.5 0.502 0.522 0.602] k 32 inertia 1077.692 cluster sizes [74 61 61 59 57 57 55 52 52 50 48 48 48 48 47 46 46 46 44 44 43 43 43 42 42 42 41 41 41 41 38 36] centroid norms [0.72 0.481 0.499 0.487 0.545 0.698 0.532 0.518 0.514 0.542 0.503 0.54 0.55 0.566 0.532 0.553 0.537 0.502 0.513 0.56 0.543 0.512 0.489 0.49 0.53 0.519 0.557 0.5 0.662 0.498 0.517 0.57 ] INDEPENDENT-COMPONENT DECOMPOSITION OF OUTPUT WEIGHT COLUMNS components 24 iterations 660 excess kurtoses [ 4.353 1.607 1.307 1.306 1.246 1.156 1.15 1.028 1.002 0.996 0.884 0.875 0.83 0.809 0.798 -0.443 -0.46 -0.493 -0.554 -0.576 -0.583 -0.591 -0.62 -0.673] absolute skewness [0.368 0.464 0.003 0.307 0.177 0.08 0.234 0.22 0.08 0.157 0.179 0.084 0.3 0.13 0.094 0.141 0.069 0.055 0.023 0.086 0.01 0.006 0.053 0.139] components 28 iterations 1416 excess kurtoses [37.202 4.452 1.572 1.357 1.232 1.232 1.154 1.153 1.131 1.131 1.083 1.049 0.934 0.925 0.876 0.854 0.795 0.789 0.779 0.745 -0.517 -0.566 -0.576 -0.58 -0.587 -0.605 -0.626 -0.702] absolute skewness [0.607 0.415 0.422 0.155 0.142 0.278 0.443 0.171 0.027 0.256 0.128 0.283 0.124 0.174 0.107 0.012 0. 0.076 0.211 0.181 0.074 0.119 0.085 0.056 0.121 0.082 0.017 0.057] INDUCTION-CANDIDATE MATRIX EIGENVALUES eigenvalues descending magnitude [ 0.0505+0.0128j 0.0505-0.0128j 0.0423+0.j 0.0356+0.0108j 0.0356-0.0108j 0.0321+0.j 0.0306+0.0061j 0.0306-0.0061j 0.0269+0.0028j 0.0269-0.0028j 0.0151+0.0061j 0.0151-0.0061j 0.0139+0.j 0.0117+0.j 0.0056+0.0101j 0.0056-0.0101j 0.0085+0.j 0.0019+0.005j 0.0019-0.005j 0.0037+0.0019j 0.0037-0.0019j -0.0019+0.0034j -0.0019-0.0034j 0.0034+0.j -0.0022+0.0025j -0.0022-0.0025j 0.0019+0.0022j 0.0019-0.0022j -0.0025+0.0002j -0.0025-0.0002j -0.0021+0.j -0.0015-0.0006j -0.0015+0.0006j 0.0008+0.0012j 0.0008-0.0012j 0.0009+0.j -0.0003+0.0007j -0.0003-0.0007j 0.0006+0.0002j 0.0006-0.0002j 0.0006+0.j 0.0002+0.0005j 0.0002-0.0005j 0.0003+0.0003j 0.0003-0.0003j -0.0002+0.0004j -0.0002-0.0004j -0.0003+0.j 0. -0.0002j 0. +0.0002j -0.0002+0.j 0.0001+0.j 0.0001-0.j -0.0001+0.j -0.0001-0.j 0.0001+0.j -0. +0.0001j -0. -0.0001j 0.0001+0.j 0.0001-0.j 0. +0.j -0. +0.j -0. +0.j -0. +0.j ] symmetric-part eigenvalues descending [ 0.0673 0.0527 0.0511 0.0468 0.0452 0.0421 0.0364 0.0329 0.0319 0.0281 0.0273 0.0188 0.0175 0.0171 0.0125 0.0103 0.0077 0.0063 0.0058 0.0044 0.0042 0.0027 0.0024 0.0021 0.0014 0.0012 0.001 0.001 0.0008 0.0004 0.0003 0.0002 0.0001 0.0001 -0.0002 -0.0003 -0.0003 -0.0005 -0.0006 -0.0007 -0.0009 -0.0012 -0.0014 -0.0017 -0.0018 -0.002 -0.0024 -0.0025 -0.0028 -0.0029 -0.0036 -0.0041 -0.0046 -0.005 -0.0052 -0.0059 -0.0073 -0.0084 -0.0086 -0.0098 -0.0103 -0.0126 -0.0142 -0.0173] ```